update package to work with SS3 version 3.30.24.2

[iantaylor-NOAA]

Inspired by an email from @JuanMatiasBraccini, I asked github copilot to update the package to work with current SS3 versions using the prompt informed by issues identified in the email:

make changes to this package to get it working with the current version of SS3 (version 3.30.24.2), available via r4ss::get_ss3_exe(). known issues include change in input filenames from ss.par to ss3.par and changes to the data.ss_new output file which is now output as 3 separate files as discussed in nmfs-ost/ss3-source-code#218"

I made some additional refinements on top of what copilot provided, including

• revising text,
• incremented the version number from 0.3.0 to 0.4.0 (but maybe it should be 0.3.1?)
• update SS3 executables from 3.30.18 to 3.30.24.2

The changes from copilot all look pretty good to me, but I've only looked over them very quickly.

@nathanvaughan-NOAA, let me know if you have any time or interest in reviewing these changes.

[iantaylor-NOAA]

I was able to work around an SS3 issue (nmfs-ost/ss3-source-code#764) by leaving the benchmarks on (which I don't think will do any harm) in c762193. However, I'm now seeing a problem in the integration test

devtools::test(filter = "interim")
i Testing SSMSE
Welcome to SSMSE version 0.4.0 using Stock Synthesis 3.30.24.2
v | F W  S  OK | Context
x | 2 1      2 | Integration test for interim assessment workflow [5.5s]            
------------------------------------------------------------------------------------
Warning (test-interim.R:33:3): run_SSMSE runs with interim assessment workflow
running command '"C:/github/SSMSE/inst/bin/Windows64/ss3.exe" -nohess' had status 1

Running the code from the test to create the failed model and then running the model outside of the test, I see the error:

Warning 4 Fatal Error! error on ss.par read; final value was not 999; total number parms changed  44.2094
Also see warning.sso
Exiting SS3!

I'm sure this won't be TOO hard to figure out, but I won't be able to get to it for at least a week. @nathanvaughan-NOAA, if you feel like debugging, feel free to fetch my branch and explore or just wait for me to get back to it in early July.

[iantaylor-NOAA]

When complete, this pull request should fix #232.
@JuanMatiasBraccini also said he can run the models with 3.30.21 and use the status-quo SSMSE as a work around.

[nathanvaughan-NOAA]

Thanks @iantaylor-NOAA I'll update you if I get a chance to continue debugging.